This application is in response to PA-09-164 (NIH Exploratory Developmental Research Grant Program). Given the high rate of hospital-acquired infection in critically ill patients, exploratory and developmental research is needed to find new methods to prevent and control infection, a major cause of intensive care unit (ICU) morbidity and mortality. An emerging body of research shows that critically ill patients have an extremely high prevalence of vitamin D insufficiency and frank vitamin D deficiency. Unfortunately, regimens of vitamin D repletion in this setting have not been explored. We propose a pilot clinical trial to test the safety and the efficacy of two high-dose vitamin D3 regimens in surgical ICU patients with lung failure. The proposed regimens are designed to increase plasma 25-hydroxyvitamin D [25(OH)D] levels into the range considered desirable by experts (> 30 ng/mL). If our strategy is efficacious, this information will be valuable to inform future investigations of vitamin D therapy and clinical outcomes in adult critical illness. Adequate vitamin D status is critical for the production of the endogenous antimicrobial peptide (AMP) LL-37 (product of cathelicidin gene) by human monocytes and epithelial cells, including from lung. Emerging data shows that vitamin D can also regulate a key member of the defensin AMP family, human beta-defensin-2 (hBD-2), a major lung AMP. Therefore, we propose to determine whether our vitamin D regimens upregulate concentrations of LL-37 and hBD-2 in plasma and bronchoalveolar lavage (BAL) fluid, cathelicidin and hBD-2 mRNA in peripheral blood mononuclear cells (PBMCs), and LL-37 and hBD-2 expression in isolated alveolar macrophages (AM). Given the role of endogenous AMPs in innate immunity and wound healing, we will generate needed data on infectious complications and related morbidities to inform subsequent study design. We hypothesize that our high-dose vitamin D regimens will: 1) safely increase plasma 25(OH)D levels into a vitamin D-sufficient range; 2) upregulate both LL-37 and hBD-2 in plasma, PBMCs, alveolar macrophages and BAL fluid; and 3) decrease hospital infection rates. Our specific aims are: 1) to perform a rigorous, double-blind, randomized, controlled, pilot clinical trial in ventilator-dependent surgical ICU patients to test the clinical/metabolic safety and efficacy of high-dose vitamin D3 therapy. The primary goal is to test whether our high-dose regimens [50,000 vs 100,000 IU D3 enterally daily for 5 consecutive days after entry (total dose = 250,000 or 500,000 units) increases plasma 25(OH)D levels to > 30 ng/mL; 2) to determine, as secondary endpoints in Aim 1 subjects, whether high-dose vitamin D3 increases LL-37 and hBD-2 levels in plasma and BAL fluid (ELISA), LL-37 and hBD-2 mRNA in PBMCs (qPCR), and LL-37 and hBD-2 protein expression in alveolar macrophages (immunohistochemistry); and 3) to test whether these vitamin D3 regimens improve alveolar macrophage phagocytosis and decrease the overall incidence of ICU- and hospital-acquired infection and associated organ dysfunction (measured as SOFA score), duration of mechanical ventilation and length of ICU and hospital stay.